[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-73420":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":8,"htmlUrl":8,"language":9,"languages":8,"totalLinesOfCode":8,"stars":10,"forks":11,"watchers":12,"openIssues":13,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":15,"stars7d":16,"stars30d":17,"stars90d":14,"forks30d":14,"starsTrendScore":16,"compositeScore":18,"rankGlobal":8,"rankLanguage":8,"license":19,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":20,"hasPages":20,"topics":22,"createdAt":8,"pushedAt":8,"updatedAt":23,"readmeContent":24,"aiSummary":25,"trendingCount":14,"starSnapshotCount":14,"syncStatus":15,"lastSyncTime":26,"discoverSource":27},73420,"amplifier","microsoft\u002Famplifier","microsoft",null,"Shell",3098,255,32,43,0,2,6,17,64.92,"MIT License",false,"main",[],"2026-06-12 04:01:09","# Amplifier\n\n**AI-powered modular development assistant - currently in early preview.**\n\n> [!CAUTION]\n> This project is a research demonstrator. It is in early development and may change significantly. Using permissive AI tools on your computer requires careful attention to security considerations and careful human supervision, and even then things can still go wrong. Use it with caution, and at your own risk, we have NOT built in the safety systems yet. We are performing our _active exploration_ in the open for others to join in the conversation and exploration, not as a product or \"official release\".\n\n> [!NOTE]\n> **Looking for the earlier Claude Code-based version?** The previous version of Amplifier, built on top of Claude Code, has been moved to the [`amplifier-claude`](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier\u002Ftree\u002Famplifier-claude) branch.\n\n---\n\n## What is Amplifier?\n\nAmplifier brings AI assistance to your command line with a modular, extensible architecture.\n\n**This CLI is _just one_ interface**—the reference implementation. The real power is the modular platform underneath. Soon you'll see web interfaces, mobile apps, voice-driven coding, and even Amplifier-to-Amplifier collaborative experiences. The community will build custom interfaces, mixing and matching modules dynamically to craft tailored AI experiences.\n\n---\n\n## Quick Start - Zero to Working in 90 Seconds\n\n> [!IMPORTANT]\n> Amplifier is currently developed and tested on macOS, Linux, and Windows Subsystem for Linux (WSL). Native Windows shells have known issues—use WSL unless you're actively contributing Windows fixes.\n\n### Step 1: Install UV (30 seconds)\n\n```bash\n# macOS\u002FLinux\u002FWSL\ncurl -LsSf https:\u002F\u002Fastral.sh\u002Fuv\u002Finstall.sh | sh\n\n# Windows PowerShell\npowershell -c \"irm https:\u002F\u002Fastral.sh\u002Fuv\u002Finstall.ps1 | iex\"\n```\n\n### Step 2: Install Amplifier (30 seconds)\n\n```bash\nuv tool install git+https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier\n```\n\n### Step 3: Run Amplifier (30 seconds)\n\n```bash\n# First-time wizard (auto-detects missing config)\namplifier init\n\n# Ask a question\namplifier run \"Explain async\u002Fawait in Python\"\n\n# Start chat mode\namplifier\n```\n\n### Step 4: Add bundles (optional)\n\n```bash\n# Add additional capability bundles\namplifier bundle add git+https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-recipes@main\namplifier bundle add git+https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-design-intelligence@main\n\n# Use a bundle\namplifier bundle use recipes\n```\n\nBundles ship focused agents you can invoke by name. Use `\u002Fagents` in chat to see available agents like `recipes:recipe-author` or `design-intelligence:component-designer`.\n\n**First time? Quick setup wizard:**\n\n\u003Cdetails>\n\u003Csummary>\u003Cb>With Anthropic Claude (recommended)\u003C\u002Fb>\u003C\u002Fsummary>\n\n```\nProvider? [1] Anthropic [2] OpenAI [3] Azure OpenAI [4] Ollama: 1\n\nAPI key: ••••••••\n  Get one: https:\u002F\u002Fconsole.anthropic.com\u002Fsettings\u002Fkeys\n✓ Saved\n\nModel? [1] claude-sonnet-4-5 [2] claude-opus-4-6 [3] custom: 1\n✓ Using claude-sonnet-4-5\n\nReady! Starting chat...\n\nReady! Starting chat...\n>\n```\n\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>\u003Cb>With Azure OpenAI (enterprise)\u003C\u002Fb>\u003C\u002Fsummary>\n\n```\nProvider? [1] Anthropic [2] OpenAI [3] Azure OpenAI [4] Ollama: 3\n\nAzure endpoint: https:\u002F\u002Fmy-resource.openai.azure.com\u002F\n✓ Saved\n\nAuthentication? [1] API key [2] Azure CLI (az login): 2\n✓ Using DefaultAzureCredential\n  (Works with 'az login' locally or managed identity in Azure)\n\nDeployment name: gpt-5.5\n  Note: Use your Azure deployment name, not model name\n✓ Configured\n\nReady! Starting chat...\n>\n```\n\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>\u003Cb>With OpenAI\u003C\u002Fb>\u003C\u002Fsummary>\n\n```\nProvider? [1] Anthropic [2] OpenAI [3] Azure OpenAI [4] Ollama: 2\n\nAPI key: ••••••••\n  Get one: https:\u002F\u002Fplatform.openai.com\u002Fapi-keys\n✓ Saved\n\nModel? [1] gpt-5.5 [2] gpt-5.4-codex [3] gpt-5.5-pro [4] custom: 1\n✓ Using gpt-5.5\n\nReady! Starting chat...\n>\n```\n\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>\u003Cb>With Ollama (local, free)\u003C\u002Fb>\u003C\u002Fsummary>\n\n```\nProvider? [1] Anthropic [2] OpenAI [3] Azure OpenAI [4] Ollama: 4\n\nModel? [1] llama3 [2] codellama [3] mistral [4] custom: 1\n✓ Using llama3\n\nMake sure Ollama is running:\n  ollama serve\n  ollama pull llama3\n\nReady! Starting chat...\n>\n```\n\n\u003C\u002Fdetails>\n\n**That's it!** From nothing to productive AI assistant in 90 seconds.\n\n## What Can Amplifier Do?\n\nFirst of all, this is still VERY early and we have not brought _most_ of our features over yet, so keep your expectations low and we'll get it ramped up very quickly over the next week or two. Consider this just an early sneak peek.\n\n- **Generate code** - From simple functions to full applications\n- **Debug problems** - Systematic error resolution with the bug-hunter agent\n- **Design systems** - Architecture planning with the zen-architect agent\n- **Research solutions** - Find patterns and best practices with the researcher agent\n- **Build modules** - Use Amplifier to create new Amplifier modules (yes, really!)\n\n**Key features:**\n\n- **Modular**: Swap AI providers, tools, and behaviors like LEGO bricks\n- **Bundle-based**: Composable configuration packages for different scenarios\n- **Session persistence**: Pick up where you left off, even across projects\n- **Extensible**: Build your own modules, bundles, or entire custom experiences\n\n**Developer Tools:**\n\n- **[Log Viewer](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-app-log-viewer)**: Web-based tool for debugging sessions with real-time log streaming and interactive JSON inspection\n\n```bash\n# Install and run the log viewer while developing\nuv tool install git+https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-app-log-viewer@main\namplifier-log-viewer\n```\n\n---\n\n## Supported AI Providers\n\nAmplifier works with multiple AI providers:\n\n- **Anthropic Claude** - Recommended, most tested (Sonnet 4.5, Opus 4.6, Haiku 4.5)\n- **OpenAI** - Good alternative (GPT-5.2, GPT-5.2-Pro, GPT-5.1-Codex)\n- **Azure OpenAI** - Enterprise users with Azure subscriptions (supports managed identity)\n- **Ollama** - Local, free, no API key needed (llama3, codellama, etc.)\n\nSwitch providers anytime:\n\n```bash\n# Switch provider (interactive - prompts for model\u002Fconfig)\namplifier provider use openai\n\n# Or explicit\namplifier provider use anthropic --model claude-opus-4-6\namplifier provider use azure-openai --deployment gpt-5.5\n```\n\n> **Note**: We've done most of our early testing with Anthropic Claude. Other providers are supported but may have rough edges we're actively smoothing out.\n\n---\n\n## Basic Usage\n\n### Interactive Chat Mode\n\n```bash\n# Start a conversation\namplifier\n\n# Or explicitly\namplifier run --mode chat\n```\n\nIn chat mode:\n\n- Context persists across messages\n- Use `\u002Fhelp` to see available commands\n- Use `\u002Ftools`, `\u002Fagents`, `\u002Fstatus`, `\u002Fconfig` to inspect session\n- Use `\u002Fthink` and `\u002Fdo` to toggle plan mode\n- Type `exit` or Ctrl+C to quit\n\n### Single Commands\n\n```bash\n# Get quick answers\namplifier run \"Explain async\u002Fawait in Python\"\n\n# Generate code\namplifier run \"Create a REST API for a todo app with FastAPI\"\n\n# Debug issues\namplifier run \"Why does this code throw a TypeError: [paste code]\"\n```\n\n### Using Bundles\n\nBundles are composable configuration packages that define tools, providers, agents, and behaviors:\n\n```bash\n# See current bundle (foundation is the default)\namplifier bundle current\n\n# List available bundles\namplifier bundle list\n\n# Use a specific bundle for one command\namplifier run --bundle recipes \"Your prompt\"\n\n# Set as default\namplifier bundle use foundation\n```\n\n**The `foundation` bundle** is the default and includes:\n\n- **Tools**: filesystem, bash, web, search, task delegation\n- **Agents**: 14 specialized agents (zen-architect, bug-hunter, git-ops, web-research, explorer, etc.)\n- **Behaviors**: logging, redaction, streaming UI, todo tracking\n\nMost users never need to change bundles—foundation provides everything for development work.\n\n### Working with Agents\n\nSpecialized agents for focused tasks:\n\n```bash\n# Let the AI delegate to specialized agents\namplifier run \"Design a caching layer with careful consideration\"\n# The AI will use zen-architect when appropriate\n\n# Or request specific agents\namplifier run \"Use bug-hunter to debug this error: [paste error]\"\n```\n\n**Bundled agents:**\n\n- **zen-architect** - System design with ruthless simplicity\n- **bug-hunter** - Systematic debugging\n- **web-research** - Web research and content fetching\n- **modular-builder** - Code implementation\n- **explorer** - Breadth-first exploration of local code, docs, and other files with citation-ready summaries\n\n---\n\n## Sessions & Persistence\n\nEvery interaction is automatically saved:\n\n```bash\n# Resume most recent session\namplifier continue\n\n# Resume with new prompt (single-shot mode)\namplifier continue \"follow-up question\"\n\n# List your recent sessions (current project only)\namplifier session list\n\n# See all sessions across all projects\namplifier session list --all-projects\n\n# View session details\namplifier session show \u003Csession-id>\n\n# Resume a specific session (interactive mode)\namplifier session resume \u003Csession-id>\n\n# Resume specific session with new prompt\namplifier run --resume \u003Csession-id> \"new question\"\n```\n\nSessions are project-scoped—when you're in `\u002Fhome\u002Fuser\u002Fmyapp`, you see only `myapp` sessions. Change directories, see different sessions. Your work stays organized.\n\n---\n\n## Configuration\n\n### Switching Providers\n\n```bash\n# Switch provider (interactive - prompts for model)\namplifier provider use openai\n\n# Or explicit\namplifier provider use anthropic --model claude-opus-4-6\n\n# Azure OpenAI (needs endpoint + deployment)\namplifier provider use azure-openai\n  Azure endpoint: https:\u002F\u002Fmy-resource.openai.azure.com\u002F\n  Auth? [1] API key [2] Azure CLI: 2\n  Deployment: gpt-5.5\n\n# Configure where to save\namplifier provider use openai --model gpt-5.5 --local      # Just you\namplifier provider use anthropic --model claude-opus-4-6 --project  # Team\n\n# See what's active\namplifier provider current\n```\n\n### Switching Bundles\n\n```bash\n# See current bundle\namplifier bundle current\n\n# Switch bundle\namplifier bundle use foundation\namplifier bundle use recipes\n\n# Add external bundles\namplifier bundle add git+https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-recipes@main\n\n# See what's active\namplifier bundle list\n```\n\n### Adding Capabilities\n\n```bash\n# Add module\namplifier module add tool-jupyter\namplifier module add tool-custom --project\n\n# See loaded modules\namplifier module current\n```\n\nSee [docs\u002FUSER_ONBOARDING.md#quick-reference](docs\u002FUSER_ONBOARDING.md#quick-reference) for complete command reference.\n\n---\n\n## Customizing Amplifier\n\n### Creating Custom Bundles\n\nBundles configure your Amplifier environment with providers, tools, agents, and behaviors.\n\n**→ [Bundle Authoring Guide](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-foundation\u002Fblob\u002Fmain\u002Fdocs\u002FBUNDLE_GUIDE.md)** - Complete guide to creating bundles\n\n### Creating Custom Agents\n\nAgents are specialized AI personas for focused tasks.\n\n**→ [Agent Authoring Guide](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-foundation\u002Fblob\u002Fmain\u002Fdocs\u002FAGENT_AUTHORING.md)** - Complete guide to creating agents\n\n---\n\n## For Developers\n\n### Building on Amplifier\n\n**Core Libraries**:\n\n- **[amplifier-core](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-core)** - Ultra-thin kernel (~2,600 lines) providing module protocols, session lifecycle, and hooks\n- **[amplifier-foundation](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-foundation)** - Bundle composition library + the default `foundation` bundle\n\n**Reference Implementation**:\n\n- **[amplifier-app-cli](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-app-cli)** - CLI application (this implementation)\n\n**Architecture**:\n\n- **[Repository Rules](docs\u002FREPOSITORY_RULES.md)** - Where docs go, what references what\n- **[Module Catalog](docs\u002FMODULES.md)** - Available providers, tools, hooks, orchestrators\n\n---\n\n## What's Next\n\n> **Note**: Amplifier is under active development. Some documentation links are being consolidated. If you encounter issues, please report them.\n\n---\n\n## The Vision\n\n**Today**: A powerful CLI for AI-assisted development.\n\n**Tomorrow**: A platform where:\n\n- **Multiple interfaces** coexist - CLI, web, mobile, voice, IDE plugins\n- **Community modules** extend capabilities infinitely\n- **Dynamic mixing** - Amplifier composes custom solutions from available modules\n- **AI builds AI** - Use Amplifier to create new modules with minimal manual coding\n- **Collaborative AI** - Amplifier instances work together on complex tasks\n\nThe modular foundation we're building today enables all of this. You're getting in early on something that's going to fundamentally change how we work with AI.\n\n---\n\n## Current State (Be Aware)\n\nThis is an **early preview release**:\n\n- APIs are stabilizing but may change\n- Some features are experimental\n- Documentation is catching up with code\n- We're moving fast—breaking changes happen\n\n**What works today:**\n\n- ✅ Core AI interactions (Anthropic Claude)\n- ✅ Bundle-based configuration\n- ✅ Agent delegation\n- ✅ Session persistence\n- ✅ Module loading from git sources\n\n**What's rough around the edges:**\n\n- ⚠️ Other providers need more testing\n- ⚠️ Some error messages could be clearer\n- ⚠️ Documentation is incomplete in places\n- ⚠️ Installation experience will improve\n\n**Having issues?** See [Troubleshooting](docs\u002FUSER_ONBOARDING.md#troubleshooting) including [Clean Reinstall](docs\u002FUSER_ONBOARDING.md#clean-reinstall-recovery) for recovery steps.\n\n**Join us on this journey!** Fork, experiment, build modules, share feedback. This is the ground floor.\n\n---\n\n## Contributing\n\n> [!NOTE]\n> This project is not currently accepting external contributions, but we're actively working toward opening this up. We value community input and look forward to collaborating in the future. For now, feel free to fork and experiment!\n\nMost contributions require you to agree to a\nContributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us\nthe rights to use your contribution. For details, visit [Contributor License Agreements](https:\u002F\u002Fcla.opensource.microsoft.com).\n\nWhen you submit a pull request, a CLA bot will automatically determine whether you need to provide\na CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions\nprovided by the bot. You will only need to do this once across all repos using our CLA.\n\nThis project has adopted the [Microsoft Open Source Code of Conduct](https:\u002F\u002Fopensource.microsoft.com\u002Fcodeofconduct\u002F).\nFor more information see the [Code of Conduct FAQ](https:\u002F\u002Fopensource.microsoft.com\u002Fcodeofconduct\u002Ffaq\u002F) or\ncontact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any additional questions or comments.\n\n## Trademarks\n\nThis project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft\ntrademarks or logos is subject to and must follow\n[Microsoft's Trademark & Brand Guidelines](https:\u002F\u002Fwww.microsoft.com\u002Flegal\u002Fintellectualproperty\u002Ftrademarks\u002Fusage\u002Fgeneral).\nUse of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship.\nAny use of third-party trademarks or logos are subject to those third-party's policies.\n","Amplifier 是一个基于 AI 的模块化开发助手，目前处于早期预览阶段。它通过命令行界面为用户提供智能辅助，具备可扩展的架构设计，允许开发者根据需求自定义功能模块，从而构建个性化的AI辅助体验。项目支持多种操作系统环境，包括macOS、Linux以及Windows Sub系统下的Linux，并且采用Shell脚本语言编写，易于安装和使用。适用于需要提高编程效率或探索AI在软件开发中应用潜力的场景，如代码解释、设计模式推荐等。由于项目仍处于研究演示阶段，在使用时需注意安全考量并谨慎操作。","2026-06-11 03:45:26","high_star"]